1

1 Attitudes toward beef and vegetarians in Argentina, Brazil, France, and the USA 2 3 Matthew B. Ruby1 4 Marle S. Alvarenga2 5 Paul Rozin1 6 Teri A. Kirby3 7 Eve Richer1 8 Guillermina Rutsztein4 9 10 1 University of Pennsylvania - USA 11 2 University of São Paulo - Brazil 12 3 University of Washington - USA 13 4 University of Buenos Aires - Argentina 14 15 16 17 Word Count (Main Body, Footnote, Acknowledgements): 6386 18 19 20 21 CORRESPONDING AUTHOR: 22 Dr. Matthew Ruby 23 University of Pennsylvania 24 3720 Walnut St, Office B13 25 Philadelphia, PA 19104 26 [email protected] 27 +1-215-898-7632 28 29 30 31 2

32 Abstract 33 Meat is both the most favored and most tabooed food in the world. In the developed world, there 34 is a tension between its high nutritional density, preferred taste, and high status on the one hand, 35 and concerns about weight, degenerative diseases, the ethics of killing animals, and the 36 environmental cost of meat production on the other hand. The present study investigated 37 attitudes toward beef, and toward vegetarians, among college students in Argentina, Brazil, 38 France, and the USA. Across countries, men were more pro-beef, in free associations, liking, 39 craving, and frequency of consumption. By country, Brazil and Argentina were generally the 40 most positive, followed by France and then the United States. Ambivalence to beef was higher in 41 women, and highest in Brazil. Only Brazilian and American women reported frequent negative 42 associations to beef (e.g. “disgusting”, “fatty”). Overall, most students had positive attitudes to 43 beef, and the attitude to vegetarians was generally neutral. America and Brazilian women 44 showed some admiration for vegetarians, while only French men and women had negative 45 attitudes to vegetarians. In spite of frequent negative ethical, health, and weight concerns, in the 46 majority of the sample, liking for and consumption of beef was maintained at a high level. 47 48 Key words: attitudes, beef, culture, gender, meat, preferences, vegetarianism 49 50 51 52 53 54 55 56 57 58 59 60 61 62 3

63 Introduction 64 In the developed world, many people have ambivalent attitudes toward meat. As a 65 concentrated source of protein, fat, and minerals, meat usually occupies a favored position in the 66 hierarchy of foods (e.g., Adams, 1990; Allen & Ng, 2003; Rozin et al., 2012; Twigg, 1979), and 67 in many societies, the ability to consume large amounts of meat has traditionally been a marker 68 of wealth and social power (e.g., Fiddes, 1991). Twigg (1979) argues that in many cultural 69 contexts, neither all food nor all meats are created equal– red meat is at the top of the pile, 70 followed by white meat, fish, dairy, eggs, and at the very bottom, fruits and vegetables. Red meat 71 is thought to occupy this position in the food hierarchy because it symbolizes power, strength, 72 and human dominance over nature through its visible blood content and associations with 73 hunting, a typically male-dominated activity (e.g., Adams, 1990; Fiddes, 1991; Sobal, 2005). 74 Although meat is held in such high regard in most societies, animal flesh is more likely 75 than vegetable matter to contain harmful bacteria and parasites (Schantz & McAuley, 1991), and 76 it is also the most frequently tabooed category of food (e.g., Fessler & Navarrete, 2003; Rozin, 77 1987; Simoons, 1994). In many Western societies, ambivalence toward eating commonly 78 consumed animals appears to be driven by several other factors. One such source of conflict (and 79 contention) is concern about the effects of high meat consumption on health. On the one hand, on 80 a country level, meat consumption is positively related to longevity, as nine of the ten most long- 81 lived countries in the world eat a diet high in meat (the exception being Japan, FAOSTAT, 82 2014), but on the other hand, within developed countries and controlling for various potential 83 confounds, such as socioeconomic status, access to healthcare, smoking, and exercise, many 84 researchers have demonstrated a relationship between meat consumption (particularly processed 85 red meat), and increased mortality, especially due to heart disease (e.g., Fraser, 1999; Huang et 86 al., 2012; Micha, Wallace, & Mozaffarian, 2010; Pan et al., 2012; Rohrmann et al., 2013). As 87 such, for many people, there is a tension between the pleasure of eating meat and concern about 88 one’s health (e.g., Becker, Kals, & Fröhlich, 2013; Povey, Wellens, & Conner, 2001; Sparks, 89 Conner, James, Shepherd, & Povey, 2001). Other major sources of concern about meat eating 90 focus on the ethics of raising and killing animals for human consumption, the environmental 91 impact of meat production, and, to a lesser but significant extent, matters of taste and expense 92 (e.g., Audebert, Deiss, & Rousset, 2006; Berndsen & van der Pligt, 2004; Richardson, Shepherd, 4

93 & Elliman, 1994; Rousset, Deiss, Juillard, Schlich, & Droit-Volet, 2005; Ruby, 2012; Ruby, 94 Heine, Kamble, Cheng, & Waddar, 2013; Singer, 1976; Steinfeld et al., 2006). 95 Presently, we are in the peculiar position that meat, a generally favored and high-status 96 human food, may be declining in popularity in the developed world because of health, 97 environmental, and ethical concerns, while its popularity is increasing in the much larger 98 developing world, as these countries become more affluent (e.g., Larsen, 2012; OECD-FAO, 99 2014). 100 Attitudes Toward Vegetarians 101 In accordance with meat ambivalence, a modest percentage of people in the world are 102 vegetarians. There is not a widely accepted definition of vegetarianism, and figures for the 103 incidence of vegetarianism are not available for many countries. Almost certainly, India has the 104 highest percentage of vegetarians, with estimates ranging from 20-42% (see Ruby, 2012 for a 105 review). In the four countries of interest in the present study, estimated prevalence of 106 vegetarianism, defined as complete avoidance of meat (that is, animal products other than dairy 107 and eggs) ranges from 8% in Brazil (Ibope, 2012), 5% in the USA (Gallup, 2012), and less than 108 2% in France (European Vegetarian Union, 2007). There is no available official estimate for 109 Argentina, the world leader for beef consumption per capita, but the Argentina Society of 110 Nutrition estimates a prevalence of 1-2% (Blanco, 2014). 111 Research on attitudes toward vegetarians is relatively scarce, and comes primarily from 112 the United States and Canada. In the first study measuring perceptions of vegetarians, Sadalla 113 and Burroughs (1986) found that US-Americans viewed vegetarians as being pacifist, 114 hypochondriacal, liberal, weight-conscious, and recreational drug-users. When asking 115 vegetarians how they saw themselves, a similar schema emerged, in that they perceived 116 themselves to be intellectual, non-competitive, weight-conscious and sexy, with a tendency to 117 use recreational drugs. More recently, in a sample of university students in the southeast United 118 States, Chin, Fisak, and Sims (2002) found that attitudes toward vegetarians were generally 119 positive, with the caveat that their sample was mostly female (81%) and liberal (65%), echoing 120 the results of Walker (1995), who found that female teenagers in the USA held more positive 121 attitudes toward vegetarians than did their male peers. 122 In yet another study conducted in the USA, Rozin, Hormes, Faith, and Wansink (2012) 123 found that participants rated targets whose favorite foods were “vegetable stir fry and other 5

124 vegetable dishes” as less masculine and more feminine than targets whose favorite foods were 125 “steak and other kinds of beef”, and in both Canadian and US-American samples, Ruby and 126 Heine (2011) found that, after controlling for perceived healthiness of diet, people perceived 127 vegetarians to be more moral and less masculine than omnivores. In another a sample of students 128 in the USA, Minson and Monin (2012) found that omnivores viewed vegetarians as virtuous but 129 weak; furthermore, the extent to which they anticipated moral reproach from vegetarians 130 predicted how much they derogated them (i.e., rating of weakness). 131 A recent set of studies by Rothgerber (2014) indicates that simply reading about 132 vegetarians can trigger a sense of guilt and dissonance in some meat eaters, leading them to 133 engage in a series of dissonance reduction strategies, such as dementalizing commonly eaten 134 animals, denying animals’ capacity for pain, and more strongly justifying their meat 135 consumption. Paradoxically, reading about a dedicated vegetarian (who never eats meat or fish, 136 as opposed to an imposter who claims to be vegetarian but frequently eats meat and fish) led 137 participants to report less frequent beef consumption, and more frequent consumption of 138 vegetarian meals– suggesting that they may have distorted their reports in an attempt to feel 139 better about their dietary choices, as simply reading a vignette could not possibly have affected 140 people’s actual past eating behavior. 141 Attitudes toward vegetarians in Brazil, Argentina, and France have not been examined, to 142 our knowledge, but on the basis of government regulations that recently required all public 143 school lunches to contain animal products, with a minimum of 20% of meals containing meat 144 and 20% containing fish, and the remainder containing egg, cheese, or offal, it appears that 145 France (French law) is unsympathetic to vegetarians (e.g., Haurant, 2011). Given the importance 146 of beef in Brazil and Argentina, and the structural opposition to people following vegetarian 147 diets in France, laypeople’s attitudes toward vegetarians in these cultural contexts remains an 148 important and underexplored topic. 149 Gender and Meat 150 In many Western societies, vegetarian women greatly outnumber vegetarian men (e.g., 151 Amato & Partridge, 1989; Fraser et al., 2000; Santos & Booth, 1996; Worsley & Skrzypiec, 152 1998) and even among Western non-vegetarians, women eat considerably less meat than men 153 (e.g., Beardsworth et al., 2002; Fraser, Welch, Luben, Binghman, & Day, 2000; Gossard & 154 York, 2003; National Public Health Institute, 1998; Richardson et al., 1993; Rimal, 2002). In 6

155 many cultural contexts, there are also large gender differences in attitudes toward meat. 156 Compared to their female peers, men in England (Beardsworth et al., 2002) and Norway (Fagerli 157 & Wandel, 1999) are more likely to believe that a healthy diet should always include meat. 158 When justifying their consumption of meat, Rothgerber (2013) found that American women 159 were more likely to use indirect methods (e.g., dissociating meat from its animal origins, 160 avoiding thinking about animal slaughter), whereas men were more likely to use direct methods 161 (e.g., claims that meat is necessary for good health, appeals to taste, human dominance over 162 nature). 163 Among a sample of American college students, Mooney and Walbourn (2001) found that 164 meat is the most commonly avoided food among female participants, and in another sample of 165 American college students, Rozin et al. (2012) found that women were more likely than men to 166 avoid eating red meat. Echoing many previous arguments about the special status of red meat 167 (e.g., Adams, 1990; Twigg, 1979; Fiddes, 1991), Rousset et al. (2005) maintain that, generally 168 speaking, “men feel hedonic pleasure in seeing and eating red meat while women experience 169 discomfort” (p. 609). Support for this statement comes from several sources. Kubberød, Ueland, 170 Rødbotten, Westad and Risvik (2002) found that among Norwegian university students, women 171 had more negative attitudes to red meat than did men, and looking at a range of different meats, 172 women disliked meats more the redder and “meatier” they were (e.g., beef, lamb). In this study, 173 and another study of Norwegian high school students (Kubberød, Ueland, Tronstad & Risvik, 174 2002), the sight of blood in red meat especially invoked images of animal death and disgust in 175 women, with similar results emerging in a sample of teenage girls in England (Kenyon & Barker, 176 1998), and in random samples of adults from the USA, UK, France, Germany, Italy, and 177 Switzerland (Ruby, Rozin, & Fischler, in prep). 178 The perceived (un)healthiness of meat appears to be another importance source of 179 ambivalence. Macht et al. (2003) found that German women were more likely to view energy- 180 dense foods as unhealthy and dangerous, and research with Danish (Holm & Møhl, 2000) and 181 South Australian (Lea & Worsley, 2002) samples reveals people’s tendency to view meat as fatty 182 and calorically dense is an important factor in their rejection. Among a sample of adult French 183 women, Audebert, Deiss, and Rousset (2006) found that enjoyment of red meat was positively 184 correlated with beliefs that meat was essential to a balanced diet and good health, and negatively 185 correlated with concern for animal breeding/slaughter practices and the impact of meat on the 7

186 environment, and Leeman, Fischler, Rozin, and Shields (2011) report more favorable attitudes to 187 the healthiness of meat in continental European than American physicians. As with general 188 attitudes toward beef, and attitudes toward vegetarianism, little is known about possible gender 189 differences in how people relate to meat in Latin America. 190 Research Questions 191 We elected to look at the place of beef in life as a function of gender and culture. We 192 selected beef because, in the Western world, it is the quintessential mammal meat. In the path to 193 vegetarianism, beef is often the first animal product to be removed from the diet (e.g., 194 Beardsworth & Keil, 1991; 1992; Santos & Booth, 1996). In particular, beef is the most 195 commonly consumed mammal meat in Argentina, Brazil, France, and the USA. Well known for 196 its asado (barbecue), Argentina is (among countries of 5 million or more people), the highest 197 consumer of beef in the world, at an estimated consumption of 54.9 kg per capita in 2011 198 (FAOSTAT, 2014). That year, Brazil’s estimate was 39.1 kg (3rd in the world), the USA 199 consumed 37.0 kg (4th in the world), and France consumed 25.4 kg (10th in the world; 200 FAOSTAT, 2014). These four countries are among the highest producers of beef– as of 2011, the 201 USA produced the most beef in the world (12.0 billion tons), followed by Brazil (#2; 9.0 billion 202 tons), Argentina (#5, 2.5 billion tons) and France (#8, 1.6 billion tons). 203 Despite its culinary and economic importance, little is known about how people relate to 204 beef in major beef-consuming countries, such as Argentina and Brazil. There is little research on 205 the psychology of beef in non-Western countries, and virtually none in South American 206 countries. Beef consumption and attitudes to beef are of particular relevance because of 207 increasing concerns in the developed world about the effects of red meat consumption on health, 208 animal welfare, and environmental sustainability. Gender is of special interest because it is well 209 documented that in the developed world, women have greater concerns about animal welfare, the 210 health effects of diet, and body weight. Beef and other red meats are often high in fat, and hence 211 high in caloric density. Thus, high liking for beef could create challenges for those concerned 212 about maintaining a lower weight. Finally, with food playing an especially important role in 213 French culture, and beef as the “central” food in Argentina, there are interesting questions about 214 how health and ethical concerns about beef impact beef attitudes in these cultures. How does 215 cultural centrality interact with the greater tendency of women, as opposed to men to be 216 concerned about animal welfare, health, and body weight? Does the centrality of beef in 8

217 Argentina and food in general in France (Rozin et al.,1999) reduce these concerns? Furthermore, 218 one might expect that countries with a greater focus on beef (or meat in general) might be more 219 hostile to those who reject these focal foods (vegetarians). In the present study, we assessed beef 220 attitudes with free associations to “beef”, liking, desire to eat, and consumption of beef, as well 221 as attitudes toward vegetarians, among university students in Brazil, Argentina, France, and the 222 USA. Given the relative lack of prior research in this domain, we conducted the study in an 223 exploratory fashion, without explicit hypotheses. 224 Method 225 Participants 226 From late 2010 to 2012, as part of a larger study on attitudes toward food, exercise, and 227 the body, we recruited a total of 1,695 participants from universities in four countries to take part 228 in “a survey on body image and attitudes toward food and physical activity.” 229 304 participants were students from the University of Buenos Aires in Argentina (84%

230 women, Mage = 23.6, SDage = 2.89; 3.0% vegetarian/vegan), and were informed of the study 231 through advertisements in psychology and anthropology classes, and via a post on the Facebook 232 page of the university’s psychology group, which was accompanied by a request for students to 233 share the post. 583 participants were students from the University of São Paulo in Brazil (62%

234 women, Mage = 21.3, SDage = 2.46; 3.9% vegetarian/vegan), and were informed of the study via 235 an email sent through the university’s academic listservs. 441 participants were students from the

236 Université de Nantes in France (62% women, Mage = 21.6, SDage = 1.46; 0.2% vegetarian/vegan), 237 and were informed of the study via an email sent to all students of the Audencia Nantes School 238 of Management. The remaining 367 participants were students from the University of

239 Pennsylvania (UPenn) in the USA (65% women, Mage = 21.5, SDage = 3.21; 5.2% 240 vegetarian/vegan), and were informed of the study via announcements in introductory and social 241 psychology courses, and via an email to all graduate students at the university. All participants 242 completed the survey on a voluntary basis; in keeping with local norms, undergraduate students 243 at UPenn received course credit for their participation, and graduate students at UPenn were 244 entered into a cash lottery with a $100 award to the winner. The US sample was intentionally 245 collected from both graduates and undergraduates for two reasons: 1) In the US, many 246 undergraduates live in dormitories, and we wanted a good representation of US students living 9

247 off campus, as is the case in the other countries; 2) US undergraduates are somewhat younger 248 than undergraduates from the other countries. 249 To guard against careless responding and to ensure more representative cross-cultural 250 comparisons, we systematically excluded data from any participants who had either left more 251 than 30% of the questionnaire blank (Argentina: 42, Brazil: 95, France: 67, USA: 13), were 252 outside the age range of 18 – 30, or who did not specify their age (Argentina: 55, Brazil: 50, 253 France: 6, USA: 43), did not specify their gender (Argentina: 19, Brazil: 1, France: 0, USA: 9), 254 or were born outside of their university’s country or had lived the majority of their life since age 255 10 outside of said country (Argentina: 5, Brazil: 2, France: 18, USA: 64). 256 Materials 257 An initial questionnaire was developed in English and pilot-tested with university 258 students in the USA. The questionnaire was then translated by native speakers of the relevant 259 languages into Brazilian Portuguese, Argentine Spanish, and French. Back translation into 260 English was done by a different set of bilingual translators, and discrepancies were resolved via 261 discussion between the translators. 262 The measures of relevance to this study assessed participant attitudes toward beef and 263 toward vegetarians, and the relation of these measures to demographic variables. Other 264 measures, not included in this report, explored body image, portion size, and general attitudes 265 toward food, eating, and exercise. The questionnaire was hosted on Surveymonkey.com. 266 To assess attitudes toward beef, participants were asked to write down the first three 267 words that come to mind when they think of beef (or carne vermelha, carne, or boeuf in 268 Brazilian Portuguese, Argentine Spanish, and French, respectively). This was the second item of 269 the questionnaire, the first being a free association to the word “chocolate”, in order to obtain 270 participants’ spontaneous responses without potential interference from other questionnaire 271 items, and without immediately indicating our focused interest on beef. After completing the free 272 associations, participants were asked to assign each word a positive, negative, or neutral value 273 (+1, 0, or -1, respectively). 274 Next, participants indicated how much they liked beef, on a scale of 0 (not at all) to 100 275 (one of your favorite foods in the world), and to indicate how often they have such a strong 276 desire for beef that they go out of their way to obtain it (1 = never, 2 = once/twice, 3 = 10

277 occasionally, 4 = often, 5 = almost daily). To assess beef consumption, participants were asked 278 to indicate how many times a month they eat beef. 279 To assess attitudes toward vegetarians, participants indicated, on a seven-point scale (1 = 280 disagree strongly, 4, neither agree nor disagree, 7 = agree strongly), their agreement/ 281 disagreement with the following items: “I admire vegetarians”, “Vegetarians bother me”, and “I 282 would prefer to date a vegetarian”. To facilitate interpretation of results, we recoded these data to 283 the metric of (-3 = disagree strongly, 0 = neither agree nor disagree, 3 = agree strongly). 284 At the end of the survey, participants provided a range of demographic data. 285 Data Analysis 286 For the qualitative data, we calculated the frequency of the words that came to mind 287 when participants were prompted to free associate to the word “beef”, using the first association 288 for each participant. We list the top 10 free associations by gender and group in Table 1, using a 289 cutoff of words that were reported by 3 or more participants. This strategy shortens the list for 290 Argentine men, which was by far the smallest group. The list represents close to raw 291 associations; we only combined clearly similar words (e.g., fat, fatty; burger, hamburger; 292 delicious, good, tasty; bad, gross, yuck, disgusting; corpse, death, dead). 293 For the quantitative data, to provide a quick overview of each set of results, and to help 294 mitigate the gender imbalance in the Argentine sample (84% women, vs. 62-65% women in the 295 other three cultures), for each measure we: 1) describe the two highest and lowest groups; 2) 296 analyze the data via a series of one-way logistic regressions (for binary outcomes) or ANOVAs 297 using Type I sums of squares (for non-binary outcomes). As our group sizes were unequal and 298 their variances were heterogeneous, to test the significance of country-level differences within 299 each gender, we used the Games-Howell post-hoc test. Finally, we examine the correlations 300 between all of our outcome variables. 301 302 Results 303 Free Associations to ‘Beef’ 304 The most frequent words (measured in terms of number of groups for which they 305 appeared in the top 10) were tasty/good (all 8 groups); cow (7 groups), fat, juicy, red, and steak 306 (5 groups). Blood, barbeque, disgusting/bad, and meat were present in 4 groups. 11

307 Most common in both Brazil and Argentina was barbecue (asado, churrasco). In both of 308 these countries, this carries the implication of socializing and sharing the meal with others. 309 The most common in France was meat (viande), followed by steak, and the most common word 310 in USA was cow, followed by meat. The French list seems most different from the others: The 311 word “disgusting/bad” was in the top 10 for all female groups except France, and references to 312 “fat/fatty” were absent only from the French list (Table 1). If we classify “fat/fatty”, 313 “death/violence”, and “disgusting/bad” as the only clearly negative words (“blood” is 314 questionable), France is the only country in which no negative words appeared in the top 10. 315 (Table 1 about here) 316 Ambivalence Toward Beef 317 We categorized participants into four mutually exclusive groups on the basis of the 318 values they assigned to each of their three free associations: 1) Ambivalent– Provided at least one 319 negative and one positive value; 2) Positive– Provided at least one positive, and NO negative 320 values; 3) Negative– Provided at least one negative, and NO positive value; 4) Neutral: Provided 321 only neutral values. We also summed the values for the three words to create a general beef 322 valence score. 323 Beef ambivalence was most common in Brazilian women (42.5%) and men (29.6%), and 324 least common in Argentine (14.0%) and American (18.6%) men. 325 We ran a series of binomial logistic regressions of gender (Woman 0 / Man 1) within 326 each culture, predicting ambivalence in spontaneous associations with beef (No = 0, Yes = 1). 327 The regressions revealed that women were more likely than men to hold ambivalent attitudes to 328 beef in Brazil [B = 0.56, Wald(1) = 9.68, p < .01], but not in Argentina, [B = 0.71, Wald(1) = 329 2.66, p = .10], France [B = 0.10, Wald(1) = 0.17, p = .68], or the USA [B = 0.30, Wald(1) = 1.19, 330 p = .28]. 331 Next, we ran a logistic regression within each gender, examining the effect of culture 332 with dummy codes for each culture. Ambivalence was more prevalent among Brazilian women 333 than among Argentine [B = 0.81, Wald(1) = 20.05, p < .001], French [B = 1.04, Wald(1) = 31.93, 334 p < .001], and American women [B = 0.88, Wald(1) = 22.12, p < .001]. The same pattern also 335 emerged among men, such that ambivalence was more prevalent among Brazilian men than 336 among Argentine [B = 0.95, Wald(1) = 4.80, p < .03], French [B = 0.57, Wald(1) = 5.46, p < .02], 337 and American men [B = 0.61, Wald(1) = 5.11, p < .03]. (See Table 2.) 12

338 Positivity Toward Beef 339 Beef positivity was most common in Argentine (82.0%) and French (71.9%) men, and 340 least common in Brazilian (40.0%) and American women (44.1%). 341 A series of logistic regressions revealed that men were more likely than women to hold 342 positive attitudes to beef in Argentina [B = 1.20, Wald(1) = 9.48, p < .01], Brazil, [B = 0.63, 343 Wald(1) = 13.37, p < .001], France [B = 0.57, Wald(1) = 7.22, p < .01], and the USA [B = 0.79, 344 Wald(1) = 12.45, p < .001] (see Table 2) 345 Positivity was more prevalent among French women than among Brazilian [B = 0.78, 346 Wald(1) = 20.49, p < .001], and American women [B = 0.61, Wald(1) = 11.41, p < .001], and 347 also more positive among Argentine women than among Brazilian [B = 0.72, Wald(1) = 18.86, p 348 < .001], and American women [B = 0.55, Wald(1) = 9.25, p < .01]. A similar pattern also 349 emerged among men, such that positivity was more prevalent among French men than Brazilian 350 [B = 0.71, Wald(1) = 10.61, p < .001], and also more positive among Argentine men than among 351 Brazilian [B = 1.29, Wald(1) = 10.85, p < .001], and American men [B = 0.96, Wald(1) = 5.45, p 352 < .02]. 353 Negativity Toward Beef 354 Beef negativity was most common in American (25.2%) and Brazilian (16.4%) women, 355 and least common in Argentine (4.0%) and French men (6.0%). 356 A series of logistic regressions revealed that women were more likely than men to hold 357 negative attitudes to beef in Argentina [B = 1.50, Wald(1) = 4.09, p < .05], France, [B = 0.99, 358 Wald(1) = 7.19, p < .01], and the USA [B = 1.39, Wald(1) = 14.77, p < .001], but not Brazil [B = 359 0.16, Wald(1) = 0.43, p = .51]. 360 Negativity was more prevalent among American women than among Argentine [B = 361 0.59, Wald(1) = 6.69, p < .01], Brazilian [B = 0.54, Wald(1) = 6.90, p < .01], and French women 362 [B = 0.68, Wald(1) = 8.94, p < .01]. Among men, negativity was more prevalent among Brazilian 363 men than French men [B = 0.97, Wald(1) = 6.55, p < .02]. (See Table 2.) 364 Neutrality Toward Beef 365 Neutral attitudes constitute the remainder after assigning positivity, negativity and 366 ambivalence, and were well below the levels for negative and positive associations, with the 367 range only 0 to 10.1% across all eight groups (see Table 2). 368 (Table 2 about here) 13

369 Valence of Associations Toward Beef 370 For the valence measure, the maximum positive score is 3.0 (3 positive associations) and 371 the minimum negative score is -3.0 (3 negative associations). Argentine (2.04) and French men 372 (1.56) were most positive toward beef, and American (0.46) and Brazilian (0.69) women were 373 least positive. 374 A two-way (Gender x Country) ANOVA revealed a significant effect of gender, such that 375 men had significantly more positive associations to beef than did women, F(1, 1687), = 42.68, p 376 < .001, d = .67. The effect of country was also significant, F(3, 1687) = 10.42, p < .001, such that 377 valence was significantly higher in Argentina and France than in the USA and Brazil. The 378 interaction of gender and country was not significant, F(3, 1687) = 1.32, p = .27. For a 379 comprehensive overview of all means and significant differences for this and all subsequent 380 quantitative measures, see Table 3. 381 (Table 3 about here) 382 Liking of Beef 383 Liking of beef (0-100 scale) was highest in French (77.88) and Argentine men (77.32), 384 and lowest in American (48.70) and Brazilian (58.27) women. 385 A two-way (Gender x Country) ANOVA revealed a significant effect of gender, such that 386 men liked beef significantly more than did women, F(1, 1682), = 133.13, p < .001, d = .59. The 387 effect of country was also significant, F(3, 1682) = 13.96, p < .001, such that liking was greatest 388 in France and Argentina, followed by Brazil, then the USA. The interaction of gender and 389 country was not significant, F(3, 1682) = 1.42, p = .23. 390 Desire to Eat Beef 391 Desire to eat beef was most frequent in French (1.96) and American (1.93) men, and least 392 frequent in Argentine (1.43) and Brazilian women (1.48). 393 There was a significant effect of gender, such that men desired beef significantly more 394 often than did women, F(1, 1687), = 31.52, p < .001, d = .35. The effect of country was also 395 significant, F(3, 1687) = 10.30, p < .001, such that desire was significantly less in Brazil and 396 Argentina than in the USA and France. The interaction of gender and country was not 397 significant, F(3, 1687) = 1.07, p = .36.1 We note a mismatch between liking, relatively high in 398 Argentines, and desire, below average in Argentines. 399 14

400 Beef Consumption 401 Frequency of beef consumption (times per month) was highest among Brazilian (25.68) 402 and Argentine (19.71) men, and lowest among American (5.97) and French (9.03) women. 403 There was a significant effect of gender, such that men consumed beef significantly more 404 often than did women, F(1, 1684), = 53.17, p < .001, d = .34. The effect of country was also 405 significant, F(3, 1684) = 123.27, p < .001, such that consumption was highest in Brazil, followed 406 by Argentina, then the USA, and finally France, with every country significantly different from 407 the other. The main effects were qualified by a significant interaction of gender and country, F(3, 408 1684) = 2.88, p < .04. Analysis of simple effects revealed that men ate beef significantly more 409 frequently than did women in Brazil [F(1, 581) = 14.08, p < .001, d = .32], France [F(1, 437) = 410 38.02, p < .001, d = .61], and the USA [F(1, 364) = 53.50, p < .001, d = .80], but not in 411 Argentina [F(1, 302), = 0.04, p = .85, d = .03]. 412 Admiration of Vegetarians 413 Admiration of vegetarians was highest in Brazilian (0.62) and American (0.56) women, 414 and lowest in French men (-1.68) and women (-1.35). Notably, only the top two groups had a 415 mean score above neutral, indicating that participants did not particularly admire vegetarians. 416 There was a significant effect of gender, such that women admired vegetarians 417 significantly more than did men, F(1, 1687), = 44.71, p < .001, d = .31. The effect of country 418 was also significant, F(3, 1687) = 124.06, p < .001, such that admiration of vegetarians was 419 highest in the USA and Brazil, followed by France, then finally Argentina. The interaction of 420 gender and country was not significant, F(3, 1687) = 1.17, p = .32. 421 Being Bothered by Vegetarians 422 American (-0.74) and French (-1.29) men were most bothered by vegetarians, and 423 Argentine women (-2.08) and men (-1.98) were least bothered. Notably, no groups had a mean 424 score above neutral, indicating that participants were not especially bothered by vegetarians. 425 There was a significant effect of gender, such that women were bothered by vegetarians 426 significantly less than were men, F(1, 1685), = 36.39, p < .001, d = .31. The effect of country 427 was also significant, F(3, 1685) = 14.38, p < .001, such that the Argentines were least bothered, 428 followed by the Brazilians and the French, and finally the Americans. The interaction of gender 429 and country was not significant, F(3, 1685) = 2.08, p = .10. 430 15

431 Aversion to Dating Vegetarians 432 American (0.05) and French (0.02) men were most averse to dating vegetarians, and 433 Argentine women (-1.53) and men (-1.20) least averse. Notably, the most averse groups were 434 just above the neutral point of the scale, indicating that participants were not especially averse to 435 dating vegetarians. 436 There was a significant effect of gender, such that women were less averse to dating 437 vegetarians than were men, F(1, 1686), = 8.55, p < .01, d = .15. The effect of country was also 438 significant, F(3, 1686) = 36.35, p < .001, such that the Argentines were least averse, followed by 439 the Brazilians, and finally the Americans and the French. The interaction of gender and country 440 was not significant, F(3, 1686) = 2.08, p = .10. 441 Relationship Between Beef and Vegetarian Measures 442 All variables (beef valence, beef liking, beef consumption, admire vegetarians, bothered 443 by vegetarians, aversion to dating vegetarians) were significantly correlated with one another at p 444 < .001, with the exception of beef consumption and aversion to dating vegetarians, which was 445 correlated at p < .05. Perhaps unsurprisingly, the strongest correlations were between beef liking 446 and the other three beef measures– beef valence (r = .56), beef desire (r = .43), and beef 447 consumption (r = .44). The vegetarian items were correlated less strongly with one another, with 448 bothered by vegetarians and aversion to dating vegetarians correlated at r = .53, but admire 449 vegetarians only correlating moderately with bothered (r = -.36) and aversion to dating (r = -.38). 450 Although all measures of attitudes toward beef were significantly correlated with attitudes 451 toward vegetarians, the strongest predictive power emerged from beef liking (admire r = -.36; 452 bothered r = .26; aversion to dating r = .31). The same general pattern emerges if one conducts 453 analyses within gender, or within country (with and without controlling for gender), although 454 correlations are higher among women than among men, and within the USA than within the 455 other 3 countries (see Table 4). 456 (Table 4 about here) 457 Discussion 458 Meat is one of the basic foods eaten by many people, and, in the Western world, beef is 459 often the most favored member of this group. At the same time, it is one of the most difficult to 460 produce foods, with considerable ethical and environmental ramifications, making it important to 461 understand how people relate to it. Beef is associated with weight gain (energy density), 16

462 increased incidence of some degenerative diseases, and ethical issues about the mistreatment and 463 killing of animals and damage to the environment. In their spontaneous free associations, 464 participants displayed both strong positive and negative attitudes toward beef. Although “tasty” 465 was in the top ten for 8/8 groups, and “juicy” in the top ten for 5/8 groups, “fat/fatty” was present 466 in 5/8, blood/bloody in 5/8, and “disgusting” in 4/8. Concern for ethical issues was far less 467 prevalent, with “death/violence” only emerging in the two Brazilian groups. Concerns about 468 eating beef, primarily regarding disgust and fat, appear more reliably among women (except in 469 France), perhaps because in France, people associate food more with pleasure than with health 470 (see Rozin, Fischler, Imada, Sarubin, & Wrzesniewski, 1999). Negativity toward beef was 471 relatively low, with the exception of American women (25.2%). Overall, it is only a minority 472 (about 25%) who express ambivalence in their thoughts about beef, although this is substantially 473 higher among Brazilian women (42%). Ambivalence may be of particular significance because 474 changes in beef attitudes, and consumption, are probably most likely in ambivalent individuals. 475 The liking of beef, a central food in all of the countries studied, is very robust– among 476 women, liking ranges from 48 to 61 (of 100), and among men, even higher (68-78). Consistent 477 gender effects emerged across all beef-related variables– relative to men, women reported fewer 478 positive free associations toward beef, held more ambivalent and fewer positive attitudes, and 479 liked and desired beef less than did men. In every country but Brazil, they held more negative 480 attitudes toward beef, and in every country but Argentina, they also ate beef less frequently than 481 did their male peers. 482 Turning to attitudes toward vegetarians, overall, women were more positively inclined 483 than were men, admiring them more in all four cultures. That said, these attitudes were 484 predominantly neutral, with only Brazilian and American women having group means above the 485 midpoint of the admiration scale. Furthermore, women in France and the USA were less 486 bothered by vegetarians than were men, and there were no significant gender differences in 487 willingness to date vegetarians. 488 At the country level, admiration of vegetarians was highest in Brazil and the USA. 489 Argentine and Brazilian participants were bothered least by vegetarians, and French participants 490 were bothered most, perhaps because, unlike the other groups, the French have a very old and 491 well-defined national cuisine that is a large part of their identity, and vegetarians could be seen 492 as a threat to this identity. That said, all groups were below the mean of the scale, indicating that 17

493 most people did not find vegetarians particularly bothersome. Similarly, Argentine and Brazilian 494 participants were least averse to dating vegetarians. 495 Our four beef variables (valence, liking, desire, and consumption) correlated positively 496 with one another in all six cases in each of the four cultures (see Table 4). Among the highest 497 correlations we report are between beef valence and beef liking (in the .50 -.58 range). These 498 high correlations suggest that the valence of free associations is a good measure of liking for 499 beef. There is evidence for some coherence of our three measures of attitudes to vegetarians, 500 although the results suggest that vegetarian admiration does not relate to any other measures in 501 France. 502 The findings we report here are just a start, and have their limitations. First, although we 503 sampled from a diverse array of cultural contexts, our sample is composed of college students, 504 who are not representative of a country as a whole. Furthermore, our sample of Argentine men is 505 rather small (N=52), so findings regarding this particular group should be interpreted with 506 caution. Finally, it is possible that the participants from France, who were students at a business 507 school, may hold more conservative views toward beef, and toward vegetarians, than students of 508 other disciplines. As such, these findings should be followed up with investigations in more 509 representative samples, in a broader array of countries, where eating beef is more taboo (e.g., 510 India, Nepal), and in countries where vegetarianism is relatively common (e.g., India, Germany), 511 and where it is virtually nonexistent (e.g., Mongolia, hunter-gatherer societies). 512 Our results suggest that there is a significant relationship between gender and attitudes 513 toward beef, with men being more positively inclined. Furthermore, there also appear to be 514 country-level differences. The Argentines and French had the most positive free associations 515 toward beef, and greatest liking for it, across genders. While the Argentines are by far the biggest 516 beef consumers, the French have the lowest intake, while being close to Argentines in liking for 517 beef. The high liking but relatively low intake in the French probably results from a number of 518 factors, including a greater role for fruits, vegetables and grains (e.g., bread) in France, and a 519 generally lower food intake (see Rozin et al., 1999). The moderation characteristic of the French 520 in the food domain seems to extend to beef intake. However, perhaps due to the pride of the 521 French in their food, the French emerge as the most hostile to vegetarians. Just as the French 522 have the least positive attitudes toward vegetarians, the American women have the most negative 523 attitudes toward beef. 18

524 The present work adds to the small but growing body of literature on how people think 525 about meat and vegetarians in different cultural settings (e.g., Beardsworth et al., 2002; Fessler & 526 Navarrete, 2003; Kubberød, Ueland, Tronstad, & Risvik, 2002; Lea & Worsley, 2002; Rousset et 527 al., 2005; Ruby, Heine, Kamble, Cheng, & Waddar, 2013). Given the shifting popularity of meat 528 (and beef in particular) in the developed and developing world, and growing concern about the 529 impact of meat consumption on issues of health, food security, and environmental sustainability 530 (e.g., Pew Commission, 2008; Steinfeld et al., 2006; Roberts, 2008), it is especially important to 531 understand attitudes toward meat and toward vegetarians. Outside of North America and 532 Western Europe, the field still knows little about these areas (for a review, see Ruby, 2012), and 533 this study helps address this knowledge gap by exploring these topics in Brazil and Argentina. 534 535 Acknowledgements 536 The authors would like to thank Mohamed Merdji for his assistance in data collection. This 537 research was supported by funds from the Positive Psychology Center at the University of 538 Pennsylvania. 539 Footnotes 540 1. Because the differences between the levels of desire to eat beef are not equal, these data are 541 technically ordinal, not ratio. Although many researchers perform ANOVAs on ordinal data, one 542 should technically use nonparametric statistics. If we instead analyzed these data with a 543 Kruskall-Wallis H Test, our inferences remain largely unchanged, save that the gender difference 544 within the Argentine sample becomes significant. 19

References Adams, C. J. (1990). The sexual politics of meat: A feminist-vegetarian critical theory. New York: Continuum. Allen, M. W., & Ng, S. H. (2003). Human values, utilitarian benefits and identification: The case of meat. European Journal of Social Psychology, 33, 37-56. Amato, P. R. & Partridge, S. A. (1989). The new vegetarians: Promoting health and protecting life. New York: Plenum Press. Audebert, O., Deiss, V., & Rousset, S. (2006). Hedonism as a predictor of attitudes of young French women towards meat. Appetite, 46, 239-47. Beardsworth, A., Bryman, A., Keil, T., Goode, J., Haslam, C., & Lancashire, E. (2002). Women, men and food: The significance of gender for nutritional attitudes and choices. British Food Journal, 104, 470-491. Beardsworth, A. D., & Keil, E. T. (1991). Vegetarianism, veganism and meat avoidance: Recent trends and findings. British Food Journal, 93(4), 19-24. Beardsworth, A., & Keil, T. (1992). The vegetarian option: Varieties, conversions, motives and careers. The Sociological Review, 40, 253-293. Becker, R., Kals, E., & Frölich, P. (2004). Meat consumption and commitments on meat policy: combining individual and public health. Journal of Health Psychology, 9, 143-155. Berndsen, M., & van der Plight, J. (2004). Ambivalence toward meat. Appetite, 42, 71-78. Bianco, D. (2014). Vegetarianos, en la mira de la ciencia: Los riesgos nutricionales de suprimir la carne. Infobae. Available online at: http://www.infobae.com/2014/04/10/1556304- vegetarianos-la-mira-la-ciencia-los-riesgos-nutricionales-suprimir-la-carne Chin, M., Fisak, B., & Sims, V. (2002). Development of the attitudes toward vegetarians scale. Anthrozoös, 15(4), 332–342. Cohen, D. (2007). Methods in cultural psychology. In S. Kitayama & D. Cohen (Eds.), Handbook of cultural psychology (pp. 196–236). New York, NY: Guilford Press. European Vegetarian Union. (2008). How Many Veggies. . .? Available online at: www.euroveg.eu/lang/en/info/howmany.php Fagerli, R. A., & Wandel, M. (1999). Gender differences in opinions and practices with regard to a "healthy diet". Appetite, 32, 171-90. FAOSTAT. (2014). Available online at: faostat3.fao.org/ 20

Fessler, D. M. T., & Navarrete, C. D. (2003). Meat is good to taboo: Dietary proscriptions as a product of the interaction of psychological mechanisms and social processes. Journal of Cognition and Culture, 3, 1-40. Fiddes, N. (1991). Meat: A natural symbol. New York: Routledge. Fox, N., & Ward, K. (2008). Health, ethics and environment: A qualitative study of vegetarian motivations. Appetite, 50, 422-429. Fraser, G. (1999). Associations between diet and cancer, ischemic heart disease, and all-cause mortality in non-Hispanic white California Seventh-day Adventists. American Journal of Clinical Nutrition, 70(suppl), 532S-538S. Fraser, G. E., Welch, A., Luben, R., Bingham, S. A., & Day, N. E. (2000). The effect of age, sex, and education on food consumption of a middle-aged english cohort— EPIC in East Anglia. Preventive Medicine, 30, 26-34. Gallup. (2012). In U.S., 5% consider themselves vegetarian. Available online at: www.gallup.com/poll/156215/consider-themselves-vegetarians.aspx Gossard, M. H., & York, R. (2003). Social structural influences on meat consumption. Human Ecology Review, 10(1), 1-9. Haurant, S. (2011). French government ‘banning vegetarianism’ in school canteens. The Guardian. Available online at: www.theguardian.com/lifeandstyle/wordofmouth/ 2011/oct/26/french-government-banning-vegetarianism-schools Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the world? Behavioral and Brain Sciences, 33, 61-135. Holm, L., & Møhl, M. (2000). The role of meat in everyday food culture. An analysis of an interview study in Copenhagen. Appetite, 34, 277–283. Huang, T., Yang, B., Li, G., Wahlqvist, M. L., & Li, D. (2012). Cardiovascular disease mortality and cancer incidence in vegetarians: A meta-analysis and systematic review. Annals of Nutrition and Metabolism, 60, 233-240. doi: 10.1159/000337301 Ibope. (2008). Dia Mundial do Vegetarianismo: 8% da população brasileira afirma ser adepta do estilo. Available online at: www.ibope.com.br/pt-br/noticias/Paginas/Dia-Mundial- do-Vegetarianismo-8-da-populacao-brasileira-afirma-ser-adepta-ao-estilo.aspx Kubberød, E., Ueland, Ø., Tronstad, A., & Risvik, E. (2002). Attitudes towards meat eating among adolescents in Norway- A qualitative study. Appetite, 38, 53–62. 21

Kubberød, E., Uelan, O., Øtten, M., Westad, F., & Risvik, E. (2002). Gender specific preferences and attitudes towards meat. Food Quality and Preference, 13(5), 285-294. Larsen, J. (2012). Peak meat: U.S. meat consumption falling. Available online at: www.earth-policy.org/data_highlights/2012/highlights25 Lea, E., & Worsley, A. (2002). The cognitive contexts of beliefs about healthiness of meat. Public Health Nutrition, 5, 37–45. Leeman, R., Fischler, C., Rozin, P., & Shields, C. (2011). Medical doctors’ attitudes and beliefs about diet and health are more like those of their lay countrymen (France, Germany, Italy, U.K. and U.S.A.) than those of doctors in other Western countries. Appetite, 56, 558-563. Macht, M., Gerer, J. & Ellgring, H. (2003). Emotions in overweight and normal-weight women immediately after eating foods differing in energy. Physiology & Behavior, 80, 367–374. Micha, R., Wallace, S. K., & Mozaffarian, D. (2010). Red and processed meat consumption and risk of incident coronary heart disease, stroke, and diabetes mellitus: A systematic review and meta-analysis. Circulation, 121, 2271-2283. Minson, J., & Monin, B. (2012). Do-gooder derogation: Trivializing morally-motivated minorities to defuse anticipated reproach. Social Personality and Psychological Science, 3, 200-207. Mooney, K. M., & Walbourn, L. (2001). When college students reject food: Not just a matter of taste. Appetite, 36, 41–50. National Public Health Institute. (1998). The 1997 dietary survey of Finnish adults. Helsinki: Publications of the National Public Health Institute B8/1998. OECD-FAO. (2014). Meats- OECD Agricultural Outlook 2014-2023. Available online at: http://stats.oecd.org/ Pan, A., Sun, Q., Bernstein, A.M., Schulze, M.B., Manson, J.E., Stampfer, M.J., … Hu, F. B. (2012).Red meat consumption and mortality: Results from 2 prospective cohort studies. Archives of Internal Medicine, 172(7), 555-63. Povey, R., Wellens, B., & Conner, M. (2001). Attitudes towards following meat, vegetarian and vegan diets: An examination of the role of ambivalence. Appetite, 37, 15–26. Richardson, N. J., Shepherd, R., & Elliman, N. A. (1993). Current attitudes and future influences on meat consumption in the U.K. Appetite, 21, 41-51. 22

Rimal, A. P. (2002). Factors affecting meat preferences among American consumers. Family Economics and Nutrition Review, 14(2), 36-43. Roberts, P. (2008).The end of food. New York, NY: Houghton Mifflin Harcourt. Rohrmann, S., Overvad, K., Bueno-de-Mesquita, H. B., Jakobsen, M. U., Egeberg, … Linseisen, J. (2013). Meat consumption and mortality - results from the European Prospective Investigation into Cancer and Nutrition. BMC Medicine, 11:63. Rothgerber, H. (2013). Real men don’t eat (vegetable) quiche: Masculinity and the justification of meat consumption. Psychology of Men & Masculinity, 14, 363-375. Rothgerber, H. (2014). Efforts to overcome vegetarian-induced dissonance among meat eaters. Appetite, 79, 32-41. Rousset, S., Deiss, V., Juillard, E., Schlich, P., & Droit-Volet, S. (2005). Emotions generated by meat and other food products in women. British Journal of Nutrition, 94(4), 609-19. Rozin, P. (1987). Psychobiological perspectives on food preferences and avoidances. In M. Harris & E.B. Ross (eds.), Food and evolution: Toward a theory of human food habits, pp. 181-205. Philadelphia: Temple University Press. Rozin, P. (2004). Meat. In Encyclopedia of Food (S. Katz, Ed.). (pp. 666-671). New York: Scribner. Rozin, P., & Alvarenga, M. (in prep). Cross-cultural differences in perceptions of appropriate portion size. Manuscript in preparation. Rozin, P., Fischler, C., Imada, S., Sarubin, A., & Wrzesniewski, A. (1999). Attitudes to food and the role of food in life in the USA, Japan, Flemish Belgium and France: Possible implications for the diet-health debate. Appetite, 33, 163-180. Rozin, P., Hormes, J. M., Faith, M. S., & Wansink, B. (2012). Is meat male? A quantitative multimethod framewok to establish metaphoric relationships. Journal of Consumer Research, 39(3), 629-643. Rozin, P., Remick, A. K., & Fischler, C. (2011). Broad themes of difference between French and Americans in attitudes to food and other life domains: Personal versus communal values, quantity versus quality, and comforts versus joys. Frontiers in Psychology, 2, 1-9. Ruby, M. B. (2012). Vegetarianism: A blossoming field of study. Appetite, 58, 141-150. Ruby, M. B., & Heine, S. J. (2011). Meat, morals, and masculinity. Appetite, 56, 447-450. 23

Ruby, M. B., Heine, S. J., Kamble, S., Cheng, T. K., & Waddar, M. (2013). Compassion and contamination: Cultural differences in vegetarianism. Appetite, 71, 340-348. Ruby, M. B., Rozin, P., & Fischler, C. (in prep). Vegetarianism and attitudes to meat across some Western, industrialized countries: Incidence, reasons and free associations to the word “meat”. Manuscript in preparation. Santos, M. L. S., & Booth, D. A. (1996). Influences on meat avoidance among British students. Appetite, 27, 197-205. Simoons, F.J. (1994). Eat not this flesh: Food avoidances from prehistory to the present. Madison: University of Wisconsin Press. Singer, P, (1976). Animal liberation. London: Jonathan Cape. Sobal, J. (2005). Men, meat, and marriage: Models of masculinity. Food and Foodways, 13, 135–158. Sparks, P., Conner, M., James, R., Sheperd, R., & Povey, R. (2001). Ambivalence about health- related behaviours: An exploration in the domain of food choice. British Journal of Health Psychology, 6, 53–68. Steinfeld, H., Gerber, P., Wassenaar, T., Castel, V., Rosales, M., & de Haan, C. (2006). Livestock’s long shadow- environmental issues and options. Food and Agriculture Organization of the United Nations: Rome, Italy. Twigg, J. (1979). Food for thought. Purity and vegetarianism. Religion, 9, 13–35. USDA. (2010). USDA agricultural projections to 2019. Washington, D. C.: Author. Available online at: www.ers.usda.gov/ersDownloadHandler.ashx?file=/media/146458/ oce101_1_.pdf Walker, C. (1995). Meet the new vegetarian. American Demographics, 17, 9–11. Wang, Y., & Beydoun, M. A. (2009). Meat consumption is associated with obesity and central obesity among US adults. International Journal of Obesity, 33(6), 621-8. Worsley, A., & Skrzypiec, G. (1998). Teenage vegetarianism: Prevalence, social and cognitive contexts. Appetite, 30, 151-170. 24

Table 1. Free associations to “beef”.

Argentina Brazil France USA Rank Women Men Women Men Women Men Women Men Asado Asado Gostosa/Saborosa/ Churrasco Viande Viande Cow Cow (BBQ) (BBQ) Bom (Tasty) (BBQ) (Meat) (Meat) 1 36 17 66 39 128 68 52 27 Rojo Rico/Gostoso Churrasco Gostosa/Saborosa/ Steak Steak Meat Steak (Red) (Tasty) (BBQ) Bom (Tasty) (Steak) (Steak) 2 31 7 64 30 23 25 24 18 Rico/Gostoso Vaca Sangue Sangue Rouge Vache Hamburger Meat (Tasty) (Cow) (Blood) (Blood) (Red) (Cow) 3 29 5 43 25 22 17 22 15 Comida Grasa/Grasosa Suculenta Proteína Vache Sang/Saignant Eww/Gross/ Hamburger (Food) (Fat/ty) (Juicy) (Protein) (Cow) (Blood/y) Yuck 4 19 3 18 17 20 8 19 7 Jugoso Jugoso Nojo/Ruim Vaca/Boi Animal Rouge Red Red (Juicy) (Juicy) (Disgust/Bad) (Cow/Ox) (Animal) (Red) 5 12 3 18 15 12 5 18 7 Animal(es) Milanesa Gordura/Gordurosa Picanha Sang/Saignant Côte Steak Protein Animal(s) (Cutlet) (Fat/Fatty) (Steak) (Blood/y) (Rib) 6 9 3 16 11 11 4 15 5 Asco Proteína Suculenta Bourguignon Délicieux/Miam Tasty/Delicious Tasty/Delicious (Disgusting) (Protein) (Juicy) (Bourgignon) (Delicious/Yum) /Good /Good 7 9 - 14 10 7 4 12 4 Nutritiva/Sana Vaca/Boi Bife Bon/Miam Bourguignon Fat/Fatty Fat/Fatty (Nutritious/Healthy) (Cow/Ox) (Beef) (Good/Yum) (Bourgignon) 8 9 - 12 5 6 3 10 4 Proteinas Picanha Gordura/Gordurosa Oeuf Gros Juicy/Saucy Jerky (Protein) (Steak) (Fat/Fatty) (Egg) (Big) 9 9 - 9 4 4 3 6 3 Sangre (Blood) 7 Morte/Violencia Morte/Violencia Hamburger Food 4 Savory (Death/Violence) (Death/Violence) (Hamburger) 10 Milanesa (Cutlet) 7 - 8 4 3 - Jerky 4 3

25

Table 2. Percent of people with ambivalent, positive, and negative free associations to “beef”.

Gender Country N Ambivalent Positive Negative Women Argentina 254 24.8 b 57.9 a * 15.7 b * Brazil 360 42.5 a * 40.0 b * 16.4 b France 274 20.8 b 59.1 a * 14.6 b * USA 367 23.5 b 44.1 b * 25.2 a * Men Argentina 50 14.0 b 82.0 a * 4.0 b * Brazil 223 29.6 a * 55.6 c * 14.3 ab France 167 19.2 b 71.9 ab * 6.0 bc * USA 129 18.6 b 63.6 bc * 7.8 b *

Note: Comparing across gender within country, percentages marked with a * differ at p < .05 or lower. Comparing across countries within gender, percentages that do not share subscripts differ at p < .05 or lower.

26

Table 3. Attitudes toward beef, beef intake, and attitudes toward vegetarians (means and standard deviations).

Beef Consumption Admire Bothered by Not Date Gender Country N Beef Valence Beef Liking Beef Desire (times/month) Vegetarians Vegetarians Vegetarians Entire Sample Argentina 304 1.24 (1.71) a 61.69 (26.31) b 1.47 (0.81) b 19.36 (14.13) b -0.22 (1.30) b -2.07 (1.45) c -1.48 (1.67) c Brazil 583 0.86 (1.86) b 63.08 (29.11) b 1.53 (0.86) b 22.61 (15.74) a 0.38 (1.87) a -1.60 (1.76) b -0.79 (2.09) b France 441 1.19 (1.55) a 67.79 (23.71) a 1.79 (0.94) a 11.30 (10.25) c -1.47 (1.61) c -1.67 (1.57) b -0.10 (2.06) a USA 367 0.74 (1.70) b 55.56 (30.04) c 1.70 (0.86) a 8.54 (9.81) d 0.30 (1.64) a -1.18 (1.58) a -0.13 (1.88) a Women Argentina 254 1.08 (1.77) 58.58 (26.29) 1.43 (0.80) 19.29 (14.83) -0.13 (1.21) -2.08 (1.45) -1.53 (1.68) Brazil 360 0.69 (1.85) 58.27 (29.51) 1.48 (0.81) 20.70 (15.21) 0.62 (1.79) -1.71 (1.68) -0.79 (2.11) France 274 0.97 (1.60) 61.64 (25.27) 1.69 (0.87) 9.03 (7.50) -1.35 (1.64) -1.91 (1.41) -0.17 (2.13) USA 367 0.46 (1.75) 48.70 (30.13) 1.58 (0.76) 5.97 (6.74) 0.56 (1.55) -1.42 (1.53) -0.22 (1.88) Men Argentina 50 2.04 (1.11) 77.32 (17.49) 1.66 (0.84) 19.71 (9.95) -0.74 (1.59) -1.98 (1.42) -1.20 (1.62) Brazil 223 1.14 (1.86) 70.80 (26.77) 1.62 (0.94) 25.68 (16.13) -0.01 (1.92) -1.43 (1.86) -0.77 (2.07) France 167 1.56 (1.40) 77.88 (16.58) 1.96 (1.02) 15.00 (12.77) -1.68 (1.55) -1.29 (1.73) 0.02 (1.96) USA 129 1.26 (1.48) 68.22 (25.49) 1.93 (0.99) 13.32 (12.51) -0.19 (1.70) -0.74 (1.58) 0.05 (1.65)

Note: For all variables, was a significant gender difference at p < .05 or lower, such that men were more positive toward meat, and less positive toward vegetarians. Country-level means that do not share subscripts differ at p < .05 or lower.

27

Table 4. Correlations between beef attitudes/consumption and attitudes toward vegetarians.

Entire Sample 1 2 3 4 5 6 7 1. Beef Valence - 2. Beef Liking .56*** - 3. Beef Desire .26*** .43*** - 4. Beef Consumption .25*** .44*** .22*** - 5. Admire Vegetarians -.28*** -.36*** -.21*** -.11*** - 6. Bothered By Vegetarians .16*** .26*** .21*** .12*** -.36*** - 7. Not Date Vegetarians .18*** .31*** .23*** .06* -.38*** .53*** - Argentina 1 2 3 4 5 6 7 1. Beef Valence - 2. Beef Liking .50*** - 3. Beef Desire .14* .36*** - 4. Beef Consumption .23*** .42*** .12* - 5. Admire Vegetarians -.09 -.22*** -.11 -.19** - 6. Bothered By Vegetarians .00 .16** .14* .13* -.25*** - 7. Not Date Vegetarians .02 .20*** .16** .08 -.17** .52*** - Brazil 1 2 3 4 5 6 7 1. Beef Valence - 2. Beef Liking .54*** - 3. Beef Desire .19*** .42*** - 4. Beef Consumption .29*** .55*** .33*** - 5. Admire Vegetarians -.32*** -.39*** -.19*** -.27*** - 6. Bothered By Vegetarians .16*** .26*** .12** .18*** -.50*** - 7. Not Date Vegetarians .21*** .34*** .20*** .20*** -.47*** .54*** - France 1 2 3 4 5 6 7 1. Beef Valence - 2. Beef Liking .53*** - 3. Beef Desire .31*** .40*** - 4. Beef Consumption .17*** .29*** .26*** - 5. Admire Vegetarians -.13** -.20*** -.12* .01 - 6. Bothered By Vegetarians .20*** .20*** .19*** .06 -.27*** - 7. Not Date Vegetarians .15** .25*** .20*** .07 -.27*** .54*** - USA 1 2 3 4 5 6 7 1. Beef Valence - 2. Beef Liking .58*** - 3. Beef Desire .38*** .47*** - 4. Beef Consumption .32*** .46*** .46*** - 5. Admire Vegetarians -.36*** -.40*** -.26*** -.28*** - 6. Bothered By Vegetarians .24*** .37*** .32*** .23*** -.54*** - 7. Not Date Vegetarians .36*** .44*** .24*** .20*** -.52*** .50*** - Note: * p < .05, * p < .01, *** p < .001. Correlations in the entire sample are standard Pearson correlation coefficients; those given separately by country partial out the effects of gender.